PHOTOVOLTAIC ENERGY HARVESTING USING MAXIMUM POWER POINT TRACKING ON A STAND ALONE SYSTEM BY Z-SOURCE INVERTER

Transcription

1 PHOTOVOLTAIC ENERGY HARVESTING USING MAXIMUM POWER POINT TRACKING ON A STAND ALONE SYSTEM BY Z-SOURCE INVERTER P.Shankar 1,Shijo james 2, Lakshmi priya.g 3 1 Assnt Prof. CSI College of Engineering, ketti. 2 Assnt Prof. CSI College of Engineering,ketti 3 Student M.E (PED) CSI College of Engineering,Ketti Abstract-- The main aim of the paper Maximum Power Point Tracking (MPPT) method for a photovoltaic energy harvesting system based on a off grid system by Z-source inverter.first it provides a brief review of Z- source inverters, MPPT methods and Model Predictive Control (MPC). Next it introduces the proposed model predictive based MPPT method. Conventionally by using hybrid system, the stability of the system is maintained and the efficiency is improved. finally a Simulink module is designed and the results are verified with the theoretical outcomes. Index Terms Impedance-Source Inverter, Model Predictive Control, Maximum Power Point Tracking, Photovoltaic Systems, Wind system. I. INTRODUCTION Natural Resources have always played a very important role in the power generation sector. Various resources such as Solar, Wind, Tidal, Geothermal, Hydro etc. contribute to the power generation sector in various ways. A large number of isolated communities, small islands, rural and village areas in developing countries are without low cost access to the electric power grid and the electricity locally produced by diesel generating groups has the economic penalty of the high cost of fuel, largely due to the added transportation costs. An Effective, economic and an efficient alternative to these diesel generating groups is the development and use of natural resources as power sources. Moreover the limited reserves of fossil fuels and global environmental concerns over their use for electric power generation have also increased the interest in the utilization of renewable energy resources. In this paper, a standalone hybrid system composed of wind turbines and photovoltaic modules is an effective method for providing electric power. Since the electric power generation in such a system is greatly affected by the weather conditions, the system must be designed and constructed to operate efficiently so as to ensure a stable and continuous electrical power supply regardless of the weather conditions. It is especially important to accurately calculate the amount of electric energy that can be generated by the wind turbine generator and the photovoltaic modules and then stored in the batteries. The Standalone system used for the study consists of Solar and Wind generation systems. However, other power generation sources such as Diesel power plants, Fuel cells etc. can also be used. The greater the system sophistication, the more suitable the power control techniques are required to be. Photovoltaic (PV) systems are one of the most promising electric power generation systems due to their low environmental impact and high availability of solar irradiation in most geographical locations [1, 2]. The energy generated by the PV systems is highly dependent on the environmental and ambient conditions such as the solar irradiance level and the module temperature. In order to ensure extraction of the maximum available energy in any ambient condition, MPPT for PV systems is essential [3]. The PV system efficiency can be degraded easily if the PV module is not forced to operate at its Maximum Power Point(MPP) at all times regardless of the environmental conditions. IJIRT INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH IN TECHNOLOGY 261

2 This paper presents a new MPPT scheme for a ZSI based PV energy harvesting system based on the concept of Model Predictive Control (MPC). The MPC technique features simplicity and flexibility, and can be programmed to compensate for the inherent non linearities associated with power electronic converters. Comparing to classical control schemes, MPC delivers a fast dynamic response with a high stability margin, making it well suited for MPPT of PV systems operating under dynamic environmental conditions. A few research works have been recently published focusing on the MPPT for grid-tied PV system by MPC. The work presented by Shadmand et al uses a conventional Perturb and Observe (P&O) algorithm for identification of MPP. However, in the approach presented in this paper, the MPC method is used directly to predict the power generated by the PV panel, subsequent to possible changes to the PV voltage. Accordingly, in this paper, the decisions on the trajectory of the PV voltage are directly made by a MPC algorithm. This provides advantages to the MPPT process over the conventional methods. Unlike the previous works, the proposed method uses a fixed switching frequency and an adaptively predicted voltage step that can change according to the proximity to the MPP. This improves the tracking response causes by variations in solar irradiance level and minimizes the oscillation around the MPP. Thus, the proposed MPPT technique features high control effectiveness, fast dynamic response, and small oscillations around MPP without requiring expensive sensing devices to measure the solar irradiance level directly. Due to nature of MPC which predicts the system behavior in a specified time horizon, the most significant advantage of the proposed technique is high accuracy tracking of gradually changing solar irradiance levels, a property absent in most wellknown MPPT techniques such as P&O. Moreover, due to small oscillations around MPP, the proposed technique makes it possible to use a ZSI with small inductors/capacitors for the PV harvesting system. This is especially important because according II. GROUNDWORK A. IMPEDANCE-SOURCE CONVERTERS The impedance networks can be utilized in a wide range of power conversion applications to provide a flexible means of conversion between different types of sources and loads [19-22]. A simple impedancesource converter, denoted as a Z- Source Inverter (ZSI) in the literature [9] is utilized as the PV harvesting interface in this paper. A distinctive characteristic of a ZSI is its capability to leverage shoot-through switching states for boosting the output voltage [11]. In shoot-through states both switches in one leg of the inverter are turned ON simultaneously. Due to inclusion of the shootthrough states, controlling ZSIs requires innovative modulation strategies. Several novel modulation strategies based on Pulse Width Modulation (PWM) method, have been proposed for ZSIs in the literature lately [9, 25, 26]. Three notable modulation strategies for ZSIs are simple-boost [11], maximum-boost [23], and constant-boost [24] techniques. In this paper, the simple boost strategy is chosen for generating the switching signals for the ZSI. The simple boost modulation strategy operates similar to a traditional carrier based PWM and it s voltage gain is given by [11], G=MB =V DC / V O/2 = M / 2M-1...(1) where M is the modulation index, B is the boosting factor of the impedance-network, ac V is the amplitude of the output voltage of the inverter (equivalent to grid peak phase voltage when gridtied), and 0 V is the dc-link voltage. The boosting factor B is given by [26], B=1/1-2D...(2) where D is the shoot through duty ratio. B. MPPT Techniques Fast convergence, small power ripple at MPP, accurate and robust tracking of MPP are the key desired properties of a MPPT technique. Several algorithms, architectures, and mechanisms for tracking the MPP of a PV module have been proposed in the literature in the past two decades. Some of the very well-known MPPT methods include: hillclimbing algorithm, power-matching scheme, curvefitting technique, P&O algorithm, incremental conductance algorithm, and fractional open-circuit IJIRT INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH IN TECHNOLOGY 262

3 voltage ( oc V ) control. In this work, the idea behind the P&O algorithm is used as grounds to develop the new model predictive based MPPT technique that features better energy harvesting efficacy can more effectively hedge against dynamic environmental conditions. algorithm tracks the MPP of the PV module by shifting the PV voltage to the voltage at MPP through the following steps, Step 1 At any given sample time k (referred to as the current sample time result, there are two possible values for the future PV Vpv (k+1) voltage at sample time k+1 (referred to as next sample time hereinafter). In the first step, the algorithm calculates the two possible future PV voltage values,... (3) Figure 1 : Block Diagram Of Proposed System III. MODEL PREDICTIVE BASED MPPT PV Side Model Predictive Based MPPT Where V is a voltage step which is an adaptively predicted value that can change according to the proximity to the MPP. In this work, the following update law for V is proposed, where is the predicted average PV voltage for the next sample time (k+1). The procedure PV Side Model Predictive Based MPPT The proposed model predictive based MPPT explained at the end of this section of finding is Step 2 In this step the algorithm calculates (predicts) the power that would be drawn from the PV module if the PV To predict the generated power, the algorithm requires the knowledge of the local P-V characteristic of the module around the operating point of Vpv (k). In this work a digital observer is designed to generate the required knowledge for the predictions. The digital observer models the PV module with the Thevenin circuit of Fig. 4. The elements of this circuit, the equivalent voltage (Veq) and equivalent resistance (Req) of the module, are functions of the P-V characteristic of the PV module and subject to local estimation by the digital observer. The employed estimator equations are, Where Vpv (k-1) and Ipv (k-1) are the values of the PV module voltage and current from the previous sampling time. Estimating the equivalent resistance and voltage of the PV module, the two possible values IJIRT INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH IN TECHNOLOGY 263

4 for the generated power in the next sampling time can be easily predicted from the next section is responsible for accomplishing this task. PROCEDURE OF FINDING PREDICTED AVERAGE...(6) Where, And... (7) In order to find the predicted average PV voltage for the next sample time, the discretized average value model of the ZSI needs to be developed. The discretized equations for the ZSI in a shoot through and a non-shoot-through state can be used to develop the average value model. The discretized equations for a non-shoot through state are found in the literature as, Where,...(8) Where T S is the sampling time and,...(10)... (11) Step 3 In this step the predicted power for the two cases will be used to evaluate the following cost function, The discretized equations for the shoot through state are found similarly from,...(9) To increase the generated power in each step, the predicted Ppv (k+1) 1 or Ppv (k+1) 2, that will result in a larger value of from (11), will be selected as the desirable trajectory for the next step. For instance, if J 1 > J 2, then the algorithm chooses to generate Ppv (k+1) 1 in the next sampling time, which correspondingly means the PV voltage will need to be shifted to Vpv (k+1) 1 by proper adjustment of the inverter gain. The desirable value of the PV voltage for the next step is denoted as Vpv (k+1) * hereinafter. In order to regulate the PV voltage to Vpv (k+1) *, the inverter gain needs to be adjusted. The ZSI power injection control system described in...(12) The V C1 (k+1) is assumed to be approximately equal to V C1 (k) since the change is minor for sufficiently small sampling time T S. The average current going through the Cpv and C 1 should be zero, thus the Ipv is the same as the ZSI inductor current I L1. Therefore the predicted average PV current can be formulated using (12) and (14) as, IJIRT INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH IN TECHNOLOGY 264

5 The proposed system uses a Proportional-Integral (PI) controller to regulate the PV and wind voltage to Vpv (k+1)* by adjusting the inverter gain. The inverter gain generated by the PI controller can be used along with the desired power factor of the operation to calculate the phase angle of the inverter voltages Фv. To calculate Фv the inverter system needs to be analyzed in a rotational q-d reference frame. The q and d axis inverter currents, I qs and I ds, can be formulated as,...(13) Considering that the relationship between the PV voltage and V C1 can be described as,... (16) the average PV voltage can be predicted... (14) where ω e,l,r,v ds,v qs,v qgs respectively represent, the grid angular frequency, line inductances, line resistances, the d axis inverter voltage, the q axis inverter voltage, and the q axis grid voltage. Additionally by substituting Where B is the boosting factor.... (15)... (17) the following equation between the inverter voltages and the inverter gain is found, B. ZSI POWER INJECTION CONTROL This part of the control system has three goals: regulating the PV voltage to Vpv(k+1)* provided by the PV side MPPT system by properly adjusting the inverter gain, controlling the ratio of power needed according to the specific application requirements, and minimizing the voltage stress on the switches. In the proposed system we have parallel wind system connected with the PV source too. The proposed control system accomplishes the three mentioned goals by generating M, D, and the phase angle of the inverter voltages Фv. The generated values will be used by the simple-boost modulator to produce proper switching signals for controlling the inverter.... (18) Moreover, the desired power factor (p.f.) can be associated with the inverter currents by the following equation... (19) Knowing the inverter gain and the power factor, (15), (17), (18) can be solved to find the inverter q and d axis voltages, V qs and V ds. Finally, by knowing V qs and V ds, the phase angle of the inverter voltages can be calculated from, IJIRT INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH IN TECHNOLOGY 265

6 ... (20) The values of M and D are generated by the voltage stress minimization. Using simple boost control, any inverter gain for a ZSI can be realized using infinite combinations of modulation indices and shootthrough duty ratios. However, inverter gains can be realized using a unique combination of M and D that will result in the minimum voltage stress on the switches. This combination can be found from, L2 0.7 mh Sampling time 60 µs Switching frequency 10 khz OUTPUT WAVEFORM For inverter gain less than or equal to one, and from, For the inverter gain more than one....(21) IV. EXPERIMENTAL RESULTS AND DISCUSSION Total Harmonic Distortion (THD) of 2.12% from these waveforms is within the IEEE-519 standards for the stand alone system. Solar irradiance level is being stepped down from 1150 W/m2 to 800W/m2. The expected pv I and pv V from the I-V characteristics of the PV modules are respectively 24.5 A and V. The actual measured values of the pv I and pv V are 20 A and V.indicating good agreement between the experimental results and the expected outcome. The voltage range has been increased. This scheme is highly for a ZSI based stand alone system and mainly used for high power applications. The experimental results demonstrate low THD which is within the IEEE 519 standards. Fast dynamic response to a step change in solar irradiance level. Negligible oscillations around MPP under dynamically changing sky condition. In future the Z source inverter can be by the quasi z- source inverter since by this the functional characteristics can be improved. Figure 2 SOLAR IRRADIANCE WAVEFORM The waveform below shows the changes in solar irradiance of the proposed technique. System Parameters PARAMETER VALUE C µf C µf L1 0.7 mh Figure 3 IJIRT INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH IN TECHNOLOGY 266

7 TOTAL HARMONIC DISTORTION REFERENCES [1] L.G. Franquelo,. S.Kouro, J.I. Leon. and D.Vinnikov., "Grid- Connected Photovoltaic Systems: An Overview of Recent Research and Emerging PV Converter Technology," IEEE Industrial Electronics Magazine, vol. 9, pp , [2] P. L. Chapman. and T. Esram, "Comparison of Photovoltaic Array Maximum Power Point Tracking Techniques," IEEE Transactions on Energy Conversion, vol. 22, pp , [3]. L. Poh Chiang,. D.M. Vilathgamuwa. and L. Yunwei., "Pulse-width modulation of Z-source inverters," IEEE Transactions on Power Electronics, vol. 20, pp , Figure 4 Total harmonic distortion has been considerably reduced. Dynamic response of the system has increased. V.CONCLUSIONS The proposed system we have combined the solar and wind energy system forming a hybrid system. The maximum power point tracking is obtained by using model predictive based method. An inverter is connected at the output of z source converter. The simulation verifies the THD level and efficiency of the system. The experimental results demonstrates low THD which is within the IEEE 519 standards. The dynamic response of the system is increased and negligible oscillations under dynamically changing sky conditions. The proposed system has applications on high power applications. [4] D.J. Adams. and Z. Peng., and, "Comparison of Traditional Inverters and Z -Source Inverter for Fuel Cell Vehicles," IEEE Transactions on Power Electronics, vol. 22, pp , [5] P. Fang Zheng., "Z-source inverter," IEEE Transactions on Industry Applications,, vol. 39, pp , [6] F. Blaabjerg., L. Poh Chiang., Y.P. Siwakoti. and G.E. Town,. "Impedance-Source Networks for Electric Power Conversion Part I: A Topological Review," IEEE Transactions on Power Electronics, vol. 30, pp , [7] H. Abu-Rub., G. Baoming., P. Fang Zheng. and L. Yushan,. "Overview of Space Vector Modulations for Three-Phase Z-Source/Quasi-Z-Source Inverters," IEEE Transactions on Power Electronics, vol. 29, pp , [8]. T. Geyer. and D. E. Quevedo,. "Multistep Finite Control Set Model Predictive Control for Power Electronics," IEEE Transactions on Power Electronics, vol. 29, pp , [9] O. Ellabban., P. Lataire. and J. Van Mierlo,. "A DSP-Based Dual-Loop Peak DC-link Voltage Control Strategy of the Z-Source Inverter," IEEE Transactions on Power Electronics, vol. 27, pp , IJIRT INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH IN TECHNOLOGY 267

8 [10] H. Jiangang., L. Jingbo. and X. Longya,. "Dynamic Modeling and Analysis of Z Source Converter-Derivation of AC Small Signal Model and Design-Oriented Analysis," IEEE Transactions on Power Electronics, vol. 22, pp , [11] H. Abu-Rub., K. Al-Haddad. and M. Malinowski Power electronics for renewable energy systems, transportation and industrial applications: John Wiley & Sons, IJIRT INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH IN TECHNOLOGY 268